AI in Higher Education: Can Machines Bring University-Level Education to the Masses?

Image title

AI in higher education

Artificial intelligence is one of the most significant pieces of technology to emerge in the 21st Century. Today, it’s never more than a few meters away — whether you’re at home watching Netflix or in your car on the way to work.

It’s hard not to think of an industry that isn’t ripe for technological innovation — especially when it comes to the delicate and vital topics of healthcare and education.

One man whose job it is to work closely with artificial intelligence solutions within the educational landscape is Dmitry Peskov.

Peskov clearly believes in a future driven by artificial intelligence, and his development of University 20.35 stands as a prime example of his ambition and eagerness to find innovative solutions for today’s Higher Education shortcomings. I sat down and asked him a couple of questions. Here are his thoughts on the future of higher education driven by AI.

You may also like: How Can AI Be Used in Schools?

Hello Dmitry, please tell us about your role within artificial intelligence-based higher education.

I am challenged with the task of training various specialists for the digital economy of the future. When we started dealing with this problem, we saw that the teaching methods and traditional educational models therein didn’t correspond to the needs of either private companies or the government. Everything is changing rapidly, new specializations are appearing, and the requirements for traditional ones are constantly expanding. We realized that we need a flexible, data-driven educational platform where everything could be personalized as much as possible through the use of AI.

How accurate can AI be in calculating students’ pathways?

It all depends on the quality of the data that we give the AI. It can take in a lot of data, but if we want to get the most useful information from it, then, of course, we should help it. For example, I already mentioned that the AI analyses various multimedia materials that students upload into the system.

But in order to draw practical conclusions from that data, it must relate to the educational process. And if you take a picture of your knee and upload it into the system, it will be just useless. Our experience shows that students quickly adapt to productive interaction with the AI. Especially when they start getting useful feedback.

Of course, in the future, the AI will require less assistance because we will gradually be able to train it to competently filter out excess information. But for now, it is like a small child that still needs to be taught a lot.

In general, the accuracy of the formation of trajectories is quite high, since the AI takes a large amount of input data for analysis. The base for it is that every digital footprint that I talked about earlier. It is clear that as the more correct data is collected, the more accurate and precise the recommended development path will be.

At the same time, the students have the opportunity to make choices on their own, as they can follow certain educational activities, choosing the ones that seem most important to them.

What sort of information is a digital footprint capable of providing about students’ capabilities?

The digital footprint includes all the student activities that we can encompass. That includes the results of tasks and achievements. The AI records how the participants interact with each other both in reality (we look at geolocation and movement) and in individual groups on social networks.

In addition, multimedia materials that the students themselves upload into the system are also analyzed. That data can include presentations, photos, videos, and audio. We also look at biometric data. In the future, I am sure that universities will definitely have to distribute loads depending on the physical and emotional state of the students because it directly affects the adoption of new materials.

This is data-driven education. And that is what will distinguish the education of the future. And it is precisely this big data that allows the AI to give the most highly personalized recommendations. During the training, a digital profile of every student is formed on the basis of their digital footprint. And the digital footprint can understand individual strengths and weaknesses more objectively than any resume.

Big Data Management Center at Island 10-22. It was here that the “digital footprint” of the participants was processed with the help of AI and the educational process was controlled online.

How does the Educational AI work in setting up an environment without the need for teachers?

It is still impossible to do without teachers and lecturers. It is just that their functions will be slightly different. In addition, the AI will allow us to objectively evaluate not only the students but also the teachers. And we are developing such mechanisms as well.

In the future, a ‘cloud’ of teachers from different universities will be formed. And, depending on the educational track, the student will receive relevant recommendations from the AI. And by the way, we are not only talking about teachers, but also individual online courses, hackathons, training, etc. And all the results will be displayed in the student’s digital profile.

Do you see Educational AI performing better than teachers in educational environments?

As I already noted, the AI does not replace lecturers or teachers. But it will allow making the transition to mass, personified education models. This is because no teacher can analyze such an amount of data. In addition, many conclusions are simply not obvious and are possible only through the use of AI.

Now there is a lot of information around us. And the problems of choice will only intensify from year to year. What kinds of skills should be developed in order for one to be in demand on the market? How and where can this be done? In this case, the AI acts as a mentor and a digital coach.

Could AI help resolve some of the big issues when it comes to education?

Yes, of course. The artificial intelligence allows for a high degree of accuracy in evaluating what is important and interesting for a particular person. A student or a schoolchild who receives only what is interesting for them learns much better. It is safe to say that the professional career of such a person will go uphill quite quickly since the person clearly knows what they need and how to achieve that. But the beginning of everything is an individual approach to education.

By the way, the analysis of speech was also carried out using the AI, and its results were recorded in the digital footprint of the participants.

The main advantage of semantic analysis is that we use it to open up a whole new level of objective assessment of individuals and in measuring the results of education. This is an assessment of the cognitive style. Neither tests, nor essays, nor any other types of exams will be able to illustrate this. An exam is, in fact, the cruelest thing. If one does not feel well, they will get an inadequate mark.

Biometric data allows to objectively assess the loads and the cognitive abilities of the participants.

Is it possible to analyze a participant’s biometrics during tasks?

We started collecting biometric data at the first intensive course at Island 10-21 in 2018. Back then, we recorded indicators using special bracelets throughout the course. This time, we decided to abandon the bracelets and organized a separate laboratory for biometrics and neurophysiology. We measured various indicators of the participants during the day and looked at how their conditions were changing. We also used various simulators and analyzed how the participants coped with various tests and tasks at the physiological level.

The collection of biometric data is necessary for greater personalization and flexibility of the educational process. Now, this is an experimental field, but in the future, the determination of loads and intensity of educational courses should be based on the state of the participants. Suppose that by the evening, we see exhaustion, then, the next day, we can adapt the program and, for example, focus on sports or cultural events.

When a person or team is in a resource state, they learn and cope with complex tasks faster and more efficiently. Therefore, we are actively working in this direction.

Further Reading

Will AI Replace Teachers?

How AI, AR, and Big Data Will Change the Future of Education

This UrIoTNews article is syndicated fromDzone

About Post Author